1. The Field of the Invention
This invention relates generally to electro-acoustic or audio loudspeaker systems. More particularly, the invention relates to a partitioning by frequency of the electrical audio signal from the output of an audio amplifier, into a plurality of frequency bands for presentation to the electro-acoustic transducers within a loudspeaker system.
2. Present State of the Art
Audio systems present as an audible signal, simultaneous divergent audio frequencies for example music or speech for appreciation by a user. The divergent frequency content of audio may generally be considered to consist of differing frequencies. While an audio system may reinforce or reproduce the electrical audio frequency spectrum in a single pair of wires or input to a speaker, specific physical implementations of speaker components are optimized for responding to a compatible band of frequencies. For example, low frequencies tend to be better replicated by physically larger drivers commonly known as woofers. Mid-range frequencies, likewise, are more favorably reproduced by a mid-range sized driver. Additionally, higher frequencies are better reproduced by physically smaller drivers commonly known as tweeters.
While an amplifier may electrically deliver the entire audio frequency spectrum to a speaker over a single pair of wires, it is impractical to expect that the high, middle and low frequencies autonomously seek out the corresponding tweeter drivers, mid-range drivers and woofer drivers within a speaker. In fact, connecting high-power, low-frequency signals to a tweeter driver, will cause audible distortion and will typically cause fatigue and destruction of the tweeter driver.
Therefore, modern higher-fidelity audio system speakers incorporate a crossover that divides the electrical audio frequency spectrum received in a single pair of wires into distinct frequency bands or ranges and ensures that only the proper frequencies are routed to the appropriate driver. That is to say, a crossover is an electric circuit or network that splits the audio frequencies into different bands for application to individual drivers. Therefore, a crossover is a key element in multiple-driver speaker system design.
Crossovers may be individually designed for a specific or custom system, or may be commercially purchased as commercial-off-the-shelf crossover networks for both two and three-way speaker systems. In a two-way speaker system, high frequencies are partitioned and routed to the tweeter driver with low frequencies being routed to the woofer driver. A two-way crossover, which uses inductors and capacitors, accomplishes this partitioning when implemented as an electrical filter. Crossover networks have heretofore incorporated at least one or more capacitors, and usually one or more inductors, and may also include one or more resistors, which are configured together to form an electrical filter for partitioning the particular audio frequencies into bands for presentation to the appropriate and compatible driver.
FIG. 1 depicts a typical two-way crossover network within a speaker system. The crossover network of FIG. 1 may be further defined as a first-order crossover network since the resultant response of each branch of the network attenuates the signal at 6 dB per octave. The graph of FIG. 1 depicts the responses of a woofer driver and a tweeter driver resulting in a first-order crossover in a two-way speaker system. An amplifier provides signal into input pair 10 comprised of a positive input 12 and a negative input 14. In the upper branch 16 of crossover network 8, the high frequencies are filtered and allowed to pass to high frequency driver 18. Filtering is performed by capacitor 20 which inhibits the passing of lower frequencies and allows the passing of higher frequencies to high frequency driver 18. Such a portion of the crossover network is commonly referred to as a "high pass" filter.
Lower frequencies are filtered through branch 22 of crossover network 8 to low frequency driver 24 through the user of the filtering element shown as inductor 26. This portion of the crossover network is commonly referred to as a "low pass" filter. It should be pointed out that crossover networks typically implement the partitioning of the frequencies into bands through the use of network branches which are parallelly configured across positive input 12 and negative input 14 of input pair 10.
The graph of FIG. 1 illustrates the frequency responses of a woofer and tweeter driver resulting from the two-way crossover network 8. Crossover network 8 is depicted as a first order crossover in a two-way speaker system. The low frequency or woofer response 28 begins rolling off at approximately 200 Hertz. As depicted in FIG. 1, at 825 Hertz, the woofer response 28 is attenuated to a negative 3 dB from the reference response of 0 dB. Tweeter response 30 is increasing in magnitude at a rate of 6 dB per octave and at 825 Hertz is also a negative 3 dB from the reference response of 0 dB. However, after 825 Hertz, tweeter response 30 increases to 0 dB while woofer response 28 continues to roll off at a rate of 6 dB per octave. The intersection of the curves depicting the woofer and tweeter response defines the "crossover frequency." Frequencies above the crossover frequency presented at input pair 10 increasingly follow the lower impedance path of branch 16 terminating at the high frequency or tweeter driver 18 rather than the higher impedance path, through branch 22, which leads to the low frequency or woofer driver 24. An implementation for selection of the crossover frequency must be carefully evaluated and selected by weighing certain characteristics to avoid further difficulties or less than ideal matching of the crossover network to the drivers of the speaker system.
FIG. 1 depicts a first-order crossover network which has a characteristic rate of attenuation of 6 dB per octave. FIG. 2 depicts a second-order crossover network which has a characteristic rate of attenuation of 12 dB per octave. FIG. 3 depicts a third-order crossover network which has a characteristic rate of attenuation of 18 dB per octave. FIG. 4 depicts a fourth-order crossover network which has a characteristic rate of attenuation of 24 dB per octave. This demonstrates that to obtain higher rates of attenuation, the number of elements in the network increases in each parallel branch of the crossover network.
Higher order crossover networks are sharper filtering devices. For example, a first order crossover network attenuates at the rate of -6 dB per octave while a second order crossover network attenuates at the rate of -12 dB per octave. Therefore, if a sufficiently low crossover frequency was selected and a first order crossover network is employed, a substantial amount of lower frequencies will still be presented to the tweeter. What this means is that such an effect causes undesirable audible distortion, limits power handling, and can easily result in tweeter damage that could be avoided by using a higher order crossover network filter.
While FIGS. 1-4 have depicted crossover networks, such examples depict that crossover networks are generally implemented as a parallel set of individual filters. Furthermore, crossover networks have heretofore required the inclusion of at least one capacitive component such as capacitor 20 for providing the requisite filtering or partitioning of the electrical audio spectrum into frequency bands. Those familiar with high-fidelity appreciate that capacitors are less than ideal components for use at speaker level signals. Furthermore, the tolerances associated with capacitors tend to lead to quite expensive component costs when attempting to accurately match or characterize components for a speaker system. Additionally, those familiar with audio systems also appreciate that the component cost, which largely includes the cost of individual components such as the capacitive components used in a crossover network, significantly affect the overall price of an audio system and in particular, the overall price associated with speakers.
Thus, what is needed is a system for partitioning the electrical audio frequency spectrum as presented by an amplifier into a plurality of frequency bands for presentment to drivers capable of reproducing the audible signal. What is yet further needed is a system for minimizing the component cost associated with an audio system, in particular speakers, through the reduction of the overall number of components required as well as through the use of more reliable and less expensive components.